Ejercicio 1¶

Rosa Huamaní Pucho (20200422)

Paso 1: Limpieza y configuración de la data¶

In [1]:
import pandas as pd
linkData="https://github.com/SocialAnalytics-StrategicIntelligence/TableOperations/raw/main/dengue_ok.pkl"

dengue = pd.read_pickle(linkData)

# checking format
dengue.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 501236 entries, 0 to 501235
Data columns (total 9 columns):
 #   Column        Non-Null Count   Dtype   
---  ------        --------------   -----   
 0   departamento  501236 non-null  object  
 1   provincia     501236 non-null  object  
 2   distrito      501236 non-null  object  
 3   ano           501236 non-null  int64   
 4   semana        501236 non-null  int64   
 5   sexo          501236 non-null  object  
 6   edad          501236 non-null  int64   
 7   enfermedad    501236 non-null  category
 8   case          501236 non-null  int64   
dtypes: category(1), int64(4), object(4)
memory usage: 31.1+ MB
In [2]:
# Each row is a person:
dengue.head()
Out[2]:
departamento provincia distrito ano semana sexo edad enfermedad case
0 HUANUCO LEONCIO PRADO LUYANDO 2000 47 M 9 SIN_SEÑALES 1
1 HUANUCO LEONCIO PRADO LUYANDO 2000 40 F 18 SIN_SEÑALES 1
2 HUANUCO LEONCIO PRADO JOSE CRESPO Y CASTILLO 2000 48 F 32 SIN_SEÑALES 1
3 HUANUCO LEONCIO PRADO JOSE CRESPO Y CASTILLO 2000 37 F 40 SIN_SEÑALES 1
4 HUANUCO LEONCIO PRADO MARIANO DAMASO BERAUN 2000 42 M 16 SIN_SEÑALES 1
In [3]:
dengue["enfermedad"].value_counts()
Out[3]:
count
enfermedad
SIN_SEÑALES 443996
ALARMA 54981
GRAVE 2259

In [4]:
dengue['enfermedad_text']=dengue.enfermedad.astype(str)

dengue.replace({'enfermedad_text':{'SIN_SEÑALES':'1_SIN_SEÑALES','ALARMA':'2_ALARMA','GRAVE':'3_GRAVE'}},inplace=True)
In [5]:
dengue_2020 =dengue["ano"] == 2020 #escogemos 2020
dengue = dengue[dengue_2020] #ahora el dataframe contiene solo datos correspondientes al 2020
In [6]:
dengue.semana.value_counts(sort=False)
Out[6]:
count
semana
1 570
5 1035
7 1095
2 702
9 1130
3 833
4 944
33 450
6 1062
11 1190
8 1060
45 1614
34 489
35 422
10 1478
50 1476
46 1616
12 1001
44 1415
43 1230
13 955
48 1455
14 1052
15 1028
16 1442
19 958
17 1271
18 1119
21 828
42 1093
20 926
31 323
22 543
23 413
49 1485
24 386
25 358
40 896
41 962
47 1665
26 280
51 1351
27 272
28 208
29 203
37 626
30 217
32 355
36 582
38 739
39 820
53 1183
52 1126

Organizamos la data por grupos de edad¶

In [7]:
binLimits=[0,15,50,110]
theLabels=["a_menor_a_16","b_entre_16y50","c_mayor_a_50"]
dengue["edad_grupos"]=pd.cut(dengue['edad'], include_lowest=True,
                                     bins=binLimits,
                                     labels=theLabels,
                                     ordered=True)

# see

dengue.head()

<ipython-input-7-f8139b7adf58>:3: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  dengue["edad_grupos"]=pd.cut(dengue['edad'], include_lowest=True,
Out[7]:
departamento provincia distrito ano semana sexo edad enfermedad case enfermedad_text edad_grupos
345299 SAN MARTIN LAMAS TABALOSOS 2020 1 F 19 SIN_SEÑALES 1 1_SIN_SEÑALES b_entre_16y50
345300 SAN MARTIN SAN MARTIN LA BANDA DE SHILCAYO 2020 1 F 19 SIN_SEÑALES 1 1_SIN_SEÑALES b_entre_16y50
345301 SAN MARTIN SAN MARTIN TARAPOTO 2020 1 M 32 SIN_SEÑALES 1 1_SIN_SEÑALES b_entre_16y50
345302 MADRE DE DIOS TAMBOPATA TAMBOPATA 2020 1 M 49 SIN_SEÑALES 1 1_SIN_SEÑALES b_entre_16y50
345303 SAN MARTIN SAN MARTIN LA BANDA DE SHILCAYO 2020 1 F 21 ALARMA 1 2_ALARMA b_entre_16y50

Paso 2: Graficamos¶

In [8]:
#!pip install --upgrade altair

Requirement already satisfied: altair in /usr/local/lib/python3.10/dist-packages (5.4.1)
Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from altair) (3.1.4)
Requirement already satisfied: jsonschema>=3.0 in /usr/local/lib/python3.10/dist-packages (from altair) (4.23.0)
Requirement already satisfied: narwhals>=1.5.2 in /usr/local/lib/python3.10/dist-packages (from altair) (1.6.0)
Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from altair) (24.1)
Requirement already satisfied: typing-extensions>=4.10.0 in /usr/local/lib/python3.10/dist-packages (from altair) (4.12.2)
Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair) (24.2.0)
Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair) (2023.12.1)
Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair) (0.35.1)
Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair) (0.20.0)
Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->altair) (2.1.5)
In [9]:
#!pip install vegafusion vegafusion-python-embed

Requirement already satisfied: vegafusion in /usr/local/lib/python3.10/dist-packages (1.6.9)
Requirement already satisfied: vegafusion-python-embed in /usr/local/lib/python3.10/dist-packages (1.6.9)
Requirement already satisfied: altair>=5.2.0 in /usr/local/lib/python3.10/dist-packages (from vegafusion) (5.4.1)
Requirement already satisfied: pyarrow>=5 in /usr/local/lib/python3.10/dist-packages (from vegafusion) (14.0.2)
Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (from vegafusion) (2.1.4)
Requirement already satisfied: psutil in /usr/local/lib/python3.10/dist-packages (from vegafusion) (5.9.5)
Requirement already satisfied: protobuf in /usr/local/lib/python3.10/dist-packages (from vegafusion) (3.20.3)
Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from altair>=5.2.0->vegafusion) (3.1.4)
Requirement already satisfied: jsonschema>=3.0 in /usr/local/lib/python3.10/dist-packages (from altair>=5.2.0->vegafusion) (4.23.0)
Requirement already satisfied: narwhals>=1.5.2 in /usr/local/lib/python3.10/dist-packages (from altair>=5.2.0->vegafusion) (1.6.0)
Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from altair>=5.2.0->vegafusion) (24.1)
Requirement already satisfied: typing-extensions>=4.10.0 in /usr/local/lib/python3.10/dist-packages (from altair>=5.2.0->vegafusion) (4.12.2)
Requirement already satisfied: numpy>=1.16.6 in /usr/local/lib/python3.10/dist-packages (from pyarrow>=5->vegafusion) (1.26.4)
Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas->vegafusion) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas->vegafusion) (2024.1)
Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas->vegafusion) (2024.1)
Requirement already satisfied: attrs>=22.2.0 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair>=5.2.0->vegafusion) (24.2.0)
Requirement already satisfied: jsonschema-specifications>=2023.03.6 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair>=5.2.0->vegafusion) (2023.12.1)
Requirement already satisfied: referencing>=0.28.4 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair>=5.2.0->vegafusion) (0.35.1)
Requirement already satisfied: rpds-py>=0.7.1 in /usr/local/lib/python3.10/dist-packages (from jsonschema>=3.0->altair>=5.2.0->vegafusion) (0.20.0)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas->vegafusion) (1.16.0)
Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->altair>=5.2.0->vegafusion) (2.1.5)
In [10]:
import pandas as pd
import altair as alt
alt.data_transformers.enable("vegafusion")
Out[10]:
DataTransformerRegistry.enable('vegafusion')

Gráfico por semanas del 2012 según intensidad de la enfermedad¶

In [11]:
#!pip install vl-convert-python>=1.6.0
In [12]:
intdengue=alt.Chart(dengue)

enc_dengue=intdengue.encode(
    x='semana:Q', #Q para datos cuanti
    y='mean(edad):Q',
    color='enfermedad_text:N', # N para datos nominales categóricos
)

enc_dengue.mark_line() + enc_dengue.mark_errorband()
Out[12]:

Gráficos según evolución de casos de dengue por semanas, sexo y grupos de edad¶

In [13]:
enc_dengue=intdengue.encode(
    x='semana:Q',
    y=alt.Y('sum(case):Q'),
    color='enfermedad_text:N',
    tooltip=['sum(case):Q','semana:Q']
).interactive()
enc_dengue.mark_line().facet(
    row='sexo:N',
    column='edad_grupos:N'
)
Out[13]:

Gráficos de barras¶

In [14]:
intdengue=alt.Chart(dengue)
enc_dengue=intdengue.encode(
    x='semana:Q',
    y=alt.Y('sum(case):Q', scale=alt.Scale(type='log')),
    column='enfermedad_text:N',
)
enc_dengue.mark_bar()
Out[14]:

Resultados en tablas¶

In [16]:
indexList =["edad_grupos","semana","sexo","enfermedad_text"]
aggregator ={'edad': ['median',"mean","min","max"]}
LevelByWeek_statsFull =dengue.groupby(indexList,observed =True).agg(aggregator)
LevelByWeek_statsFull
Out[16]:
edad
median mean min max
edad_grupos semana sexo enfermedad_text
a_menor_a_16 1 F 1_SIN_SEÑALES 10.0 9.064935 1 15
2_ALARMA 9.5 9.428571 2 15
3_GRAVE 3.0 3.000000 2 4
M 1_SIN_SEÑALES 10.0 9.736842 1 15
2_ALARMA 11.5 10.444444 4 15
... ... ... ... ... ... ... ...
c_mayor_a_50 52 M 2_ALARMA 64.0 63.909091 53 82
53 F 1_SIN_SEÑALES 58.0 61.458333 51 92
2_ALARMA 63.0 65.000000 52 87
M 1_SIN_SEÑALES 58.0 62.898305 51 90
2_ALARMA 60.0 61.000000 51 78

752 rows × 4 columns

In [17]:
LevelByWeek_statsFull.stack(future_stack=True)
Out[17]:
edad
edad_grupos semana sexo enfermedad_text
a_menor_a_16 1 F 1_SIN_SEÑALES median 10.000000
mean 9.064935
min 1.000000
max 15.000000
2_ALARMA median 9.500000
... ... ... ... ... ...
c_mayor_a_50 53 M 1_SIN_SEÑALES max 90.000000
2_ALARMA median 60.000000
mean 61.000000
min 51.000000
max 78.000000

3008 rows × 1 columns

In [18]:
indexList=['semana','departamento','provincia','enfermedad_text']
aggregator={'case':['sum']}
ByWeekPlace=dengue.groupby(indexList,observed=True).agg(aggregator)
ByWeekPlace
Out[18]:
case
sum
semana departamento provincia enfermedad_text
1 AMAZONAS BAGUA 2_ALARMA 1
CUSCO LA CONVENCION 1_SIN_SEÑALES 6
QUISPICANCHI 1_SIN_SEÑALES 6
JUNIN CHANCHAMAYO 1_SIN_SEÑALES 11
2_ALARMA 3
... ... ... ... ...
53 UCAYALI ATALAYA 1_SIN_SEÑALES 3
CORONEL PORTILLO 1_SIN_SEÑALES 243
2_ALARMA 27
PADRE ABAD 1_SIN_SEÑALES 14
2_ALARMA 1

2842 rows × 1 columns

In [21]:
# no missing values
ByWeekPlace_wide =ByWeekPlace.unstack().fillna(0)
ByWeekPlace_wide
Out[21]:
case
sum
enfermedad_text 1_SIN_SEÑALES 2_ALARMA 3_GRAVE
semana departamento provincia
1 AMAZONAS BAGUA 0.0 1.0 0.0
CUSCO LA CONVENCION 6.0 0.0 0.0
QUISPICANCHI 6.0 0.0 0.0
JUNIN CHANCHAMAYO 11.0 3.0 0.0
SATIPO 9.0 1.0 0.0
... ... ... ... ... ...
53 TUMBES TUMBES 20.0 0.0 0.0
ZARUMILLA 1.0 0.0 0.0
UCAYALI ATALAYA 3.0 0.0 0.0
CORONEL PORTILLO 243.0 27.0 0.0
PADRE ABAD 14.0 1.0 0.0

1872 rows × 3 columns

Proporción de personas con alarma por dengue¶

In [22]:
sumCases=ByWeekPlace_wide.sum(axis=1)
sumCases
Out[22]:
0
semana departamento provincia
1 AMAZONAS BAGUA 1.0
CUSCO LA CONVENCION 6.0
QUISPICANCHI 6.0
JUNIN CHANCHAMAYO 14.0
SATIPO 10.0
... ... ... ...
53 TUMBES TUMBES 20.0
ZARUMILLA 1.0
UCAYALI ATALAYA 3.0
CORONEL PORTILLO 270.0
PADRE ABAD 15.0

1872 rows × 1 columns


In [23]:
# here you are:
shareAlarma=ByWeekPlace_wide.loc[:,('case','sum','2_ALARMA')]/sumCases
shareAlarma.name='shareAlarma'
shareAlarma
Out[23]:
shareAlarma
semana departamento provincia
1 AMAZONAS BAGUA 1.000000
CUSCO LA CONVENCION 0.000000
QUISPICANCHI 0.000000
JUNIN CHANCHAMAYO 0.214286
SATIPO 0.100000
... ... ... ...
53 TUMBES TUMBES 0.000000
ZARUMILLA 0.000000
UCAYALI ATALAYA 0.000000
CORONEL PORTILLO 0.100000
PADRE ABAD 0.066667

1872 rows × 1 columns


In [24]:
shareAlarma=shareAlarma.reset_index()
shareAlarma
Out[24]:
semana departamento provincia shareAlarma
0 1 AMAZONAS BAGUA 1.000000
1 1 CUSCO LA CONVENCION 0.000000
2 1 CUSCO QUISPICANCHI 0.000000
3 1 JUNIN CHANCHAMAYO 0.214286
4 1 JUNIN SATIPO 0.100000
... ... ... ... ...
1867 53 TUMBES TUMBES 0.000000
1868 53 TUMBES ZARUMILLA 0.000000
1869 53 UCAYALI ATALAYA 0.000000
1870 53 UCAYALI CORONEL PORTILLO 0.100000
1871 53 UCAYALI PADRE ABAD 0.066667

1872 rows × 4 columns

Encontremos la peor provincia por Región en las semanas del 2020

In [25]:
where = shareAlarma.groupby(['semana','departamento'])['shareAlarma'].idxmax()
worst_prov_Week = shareAlarma.loc[where].reset_index(drop=True)
worst_prov_Week
Out[25]:
semana departamento provincia shareAlarma
0 1 AMAZONAS BAGUA 1.000000
1 1 CUSCO LA CONVENCION 0.000000
2 1 JUNIN CHANCHAMAYO 0.214286
3 1 LORETO MARISCAL RAMON CASTILLA 0.800000
4 1 MADRE DE DIOS TAMBOPATA 0.294686
... ... ... ... ...
777 53 MADRE DE DIOS TAHUAMANU 0.400000
778 53 PASCO OXAPAMPA 0.000000
779 53 SAN MARTIN EL DORADO 0.500000
780 53 TUMBES CONTRALMIRANTE VILLAR 0.000000
781 53 UCAYALI CORONEL PORTILLO 0.100000

782 rows × 4 columns

In [27]:
len(worst_prov_Week.provincia.value_counts())
Out[27]:
64
In [29]:
len(worst_prov_Week[worst_prov_Week.shareAlarma>0].provincia.value_counts()) # amount of worst provinces per region - cleaner
Out[29]:
58

Aplicamos algunos filtros:

In [30]:
worst_ProvWeek_alarma=worst_prov_Week[worst_prov_Week.shareAlarma>0].loc[:,['departamento','provincia']]
worst_ProvWeek_alarma.reset_index(drop=True,inplace=True)
worst_ProvWeek_alarma
Out[30]:
departamento provincia
0 AMAZONAS BAGUA
1 JUNIN CHANCHAMAYO
2 LORETO MARISCAL RAMON CASTILLA
3 MADRE DE DIOS TAMBOPATA
4 SAN MARTIN HUALLAGA
... ... ...
464 JUNIN SATIPO
465 LORETO MARISCAL RAMON CASTILLA
466 MADRE DE DIOS TAHUAMANU
467 SAN MARTIN EL DORADO
468 UCAYALI CORONEL PORTILLO

469 rows × 2 columns

In [31]:
indexList=['departamento','provincia']
aggregator={'provincia':['count']}
worst_ProvWeek_alarma_Frequency=worst_ProvWeek_alarma.groupby(indexList,observed=True).agg(aggregator)
worst_ProvWeek_alarma_Frequency
Out[31]:
provincia
count
departamento provincia
AMAZONAS BAGUA 10
CONDORCANQUI 12
UTCUBAMBA 10
AYACUCHO CANGALLO 1
HUANTA 7
LA MAR 7
PARINACOCHAS 1
CAJAMARCA JAEN 22
CALLAO CALLAO 1
CUSCO LA CONVENCION 28
HUANUCO HUAMALIES 1
HUANUCO 1
LEONCIO PRADO 11
PUERTO INCA 10
ICA CHINCHA 1
ICA 18
NAZCA 2
PISCO 6
JUNIN CHANCHAMAYO 19
HUANCAYO 1
SATIPO 31
LA LIBERTAD ASCOPE 4
PACASMAYO 1
TRUJILLO 2
LAMBAYEQUE CHICLAYO 2
LAMBAYEQUE 1
LIMA LIMA 14
LORETO ALTO AMAZONAS 7
DATEM DEL MARAÑON 1
LORETO 8
MARISCAL RAMON CASTILLA 9
MAYNAS 18
PUTUMAYO 3
REQUENA 3
UCAYALI 1
MADRE DE DIOS MANU 6
TAHUAMANU 5
TAMBOPATA 31
PASCO OXAPAMPA 11
PIURA PAITA 1
PIURA 9
SULLANA 3
PUNO CARABAYA 1
SAN MARTIN BELLAVISTA 1
EL DORADO 3
HUALLAGA 11
LAMAS 2
MARISCAL CACERES 6
MOYOBAMBA 3
PICOTA 1
RIOJA 2
SAN MARTIN 14
TOCACHE 4
TUMBES TUMBES 22
ZARUMILLA 6
UCAYALI ATALAYA 11
CORONEL PORTILLO 24
PADRE ABAD 18
In [32]:
#limpieza final
worst_ProvWeek_alarma_Frequency.columns=['WeekAffected']
worst_ProvWeek_alarma_Frequency=worst_ProvWeek_alarma_Frequency[worst_ProvWeek_alarma_Frequency.WeekAffected>2]
worst_ProvWeek_alarma_Frequency.reset_index(inplace=True)
worst_ProvWeek_alarma_Frequency
Out[32]:
departamento provincia WeekAffected
0 AMAZONAS BAGUA 10
1 AMAZONAS CONDORCANQUI 12
2 AMAZONAS UTCUBAMBA 10
3 AYACUCHO HUANTA 7
4 AYACUCHO LA MAR 7
5 CAJAMARCA JAEN 22
6 CUSCO LA CONVENCION 28
7 HUANUCO LEONCIO PRADO 11
8 HUANUCO PUERTO INCA 10
9 ICA ICA 18
10 ICA PISCO 6
11 JUNIN CHANCHAMAYO 19
12 JUNIN SATIPO 31
13 LA LIBERTAD ASCOPE 4
14 LIMA LIMA 14
15 LORETO ALTO AMAZONAS 7
16 LORETO LORETO 8
17 LORETO MARISCAL RAMON CASTILLA 9
18 LORETO MAYNAS 18
19 LORETO PUTUMAYO 3
20 LORETO REQUENA 3
21 MADRE DE DIOS MANU 6
22 MADRE DE DIOS TAHUAMANU 5
23 MADRE DE DIOS TAMBOPATA 31
24 PASCO OXAPAMPA 11
25 PIURA PIURA 9
26 PIURA SULLANA 3
27 SAN MARTIN EL DORADO 3
28 SAN MARTIN HUALLAGA 11
29 SAN MARTIN MARISCAL CACERES 6
30 SAN MARTIN MOYOBAMBA 3
31 SAN MARTIN SAN MARTIN 14
32 SAN MARTIN TOCACHE 4
33 TUMBES TUMBES 22
34 TUMBES ZARUMILLA 6
35 UCAYALI ATALAYA 11
36 UCAYALI CORONEL PORTILLO 24
37 UCAYALI PADRE ABAD 18

Gráfico de afectados por semana de acuerdos a la provincia y departamentos¶

In [33]:
alt_worstProv=alt.Chart(worst_ProvWeek_alarma_Frequency)

enc_worstProv=alt_worstProv.encode(
    y='departamento',
    x='provincia',
    text='WeekAffected:O',
    size='WeekAffected:O'
)

enc_worstProv.mark_text()
Out[33]:
In [34]:
indexList=['semana','provincia','enfermedad_text']
aggregator={'case':['sum']}
ByWeekProv=dengue.groupby(indexList,observed=True).agg(aggregator)
ByWeekProv_wide=ByWeekProv.unstack().fillna(0)
ByWeekProvAlarm=ByWeekProv_wide.loc[:,('case','sum','2_ALARMA')]/ByWeekProv_wide.sum(axis=1)
ByWeekProvAlarm.name='alarmShare'

ByWeekProvAlarm=ByWeekProvAlarm.reset_index()
ByWeekProvAlarm
Out[34]:
semana provincia alarmShare
0 1 ALTO AMAZONAS 0.285714
1 1 BAGUA 1.000000
2 1 BELLAVISTA 0.000000
3 1 CARABAYA 0.000000
4 1 CHANCHAMAYO 0.214286
... ... ... ...
1867 53 TOCACHE 0.125000
1868 53 TUMBES 0.000000
1869 53 UCAYALI 0.140000
1870 53 UTCUBAMBA 0.052632
1871 53 ZARUMILLA 0.000000

1872 rows × 3 columns

In [35]:
ByWeekProvAlarm_focus=ByWeekProvAlarm[ByWeekProvAlarm.alarmShare>0]
In [39]:
ByWeekProvAlarm_focus.describe()
Out[39]:
semana alarmShare
count 893.000000 893.000000
mean 28.017917 0.278162
std 16.756824 0.267728
min 1.000000 0.001248
25% 12.000000 0.090909
50% 29.000000 0.190476
75% 44.000000 0.333333
max 53.000000 1.000000
In [40]:
edges=[-1, .10, .25, .5,1]
theLabels=["a.below10%","b.11-25%","c.26-50%","d.above50%"]
ByWeekProvAlarm_focus.loc[:,"alarmLevels"]=pd.cut(ByWeekProvAlarm_focus['alarmShare'],
                                            include_lowest=True,
                                            bins=edges,
                                            labels=theLabels,
                                            ordered=True)

##
ByWeekProvAlarm_focus.head()
Out[40]:
semana provincia alarmShare alarmLevels
0 1 ALTO AMAZONAS 0.285714 c.26-50%
1 1 BAGUA 1.000000 d.above50%
4 1 CHANCHAMAYO 0.214286 b.11-25%
5 1 CORONEL PORTILLO 0.222222 b.11-25%
7 1 EL DORADO 0.333333 c.26-50%
In [44]:
alt_WorstProv=alt.Chart(ByWeekProvAlarm_focus).encode(x='semana:O',
                                                      y=alt.Y('provincia:N',
                                                              sort=alt.EncodingSortField(field='alarmShare',op='max',order='descending')))
enc1_WorstProv=alt_WorstProv.encode(
    color=alt.Color('alarmLevels:O').scale(scheme="lightgreyred", reverse=False)
)
enc1_WorstProv.mark_rect()
Out[44]:
In [45]:
enc2_WorstProv=alt_WorstProv.encode(
    text=alt.Text('alarmShare:Q', format=".1f"),
    opacity=alt.condition('datum.alarmShare >= 0.3', alt.value(1), alt.value(0)))
enc2_WorstProv.mark_text(fontStyle='bold')
Out[45]:

Graficamos las semanas en alerta por provincias¶

In [46]:
enc1_WorstProv.mark_rect() + enc2_WorstProv.mark_text()
Out[46]:
In [ ]:
In [ ]:
!pip install nbconvert
!jupyter nbconvert --to HTML "tarea1_20200422(1)".ipynb